No op named DecodeWav in defined operations.

[arm-software]

No op named DecodeWav in defined operations.

[navsuda]

@Shahnawax , it seems you are using an older version of tensorflow. Please use TF 1.4 for loading the trained models.

[Shahnawax]

Dear Naveen, thank you for quick reply.

I tried tensorflow version 1.0.1, 1.4.0 and 1.4.1. With all versions I am getting the same error. :(

[navsuda]

Hi @Shahnawax, it may be related to this: tensorflow/tensorflow#12657. Also, can you try generating the .pb file on the same machine as you are testing it. (Training shouldn't take very long, ~1hr on a decent GPU).

[NanWang0024]

To confirm, tensorflow/tensorflow#12657 successfully solved this problem.

A side note: better to use TF 1.5+ since another error (No module named 'tensorflow.python.ops.gen_audio_ops') showed up in 1.4 as described in tensorflow/tensorflow#14004.

Successfully loaded the models with tf_nightly-1.6.0.dev20180126-cp36-cp36m-win_amd64.whl.

[navsuda]

Thanks @NanWang0024 .
Closing the issue as it is fixed in TF 1.5.

[Shahnawax]

Thank you :)

[ccnankai]

@Shahnawax
In the quantization process, how to get the value of the quantization ranges -- act_max in quant_test.py .
how to set the specific value of the parameter act_max ( x 0 0 0 0 0 0 0 0 0 0 0 ) ?